Professional Documents
Culture Documents
Wearables and The Internet of Things IoT Applications Opportunities and Challenges A Survey
Wearables and The Internet of Things IoT Applications Opportunities and Challenges A Survey
Wearables and The Internet of Things IoT Applications Opportunities and Challenges A Survey
23, 2020.
Digital Object Identifier 10.1109/ACCESS.2020.2986329
ABSTRACT Smart wearables collect and analyze data, and in some scenarios make a smart decision
and provide a response to the user and are finding more and more applications in our daily life. In this
paper, we comprehensively survey the most recent and important research works conducted in the area of
wearable Internet of Things (IoT) and classify the wearables into four major clusters: (i) health, (ii) sports and
daily activity, (iii) tracking and localization, and (iv) safety. The fundamental differences of the algorithms
associated within each cluster are grouped and analyzed and the research challenges and open issues in each
cluster are discussed. This survey reveals that although Cellular IoT (CIoT) has many advantages and can
bring enormous applications to IoT wearables, it has been rarely studied by the researchers. This article also
addresses the opportunities and challenges related to implementing CIoT-enabled wearables.
This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/
69200 VOLUME 8, 2020
F. J. Dian et al.: Wearables and the IoT, Applications, Opportunities, and Challenges
where the sensor is implemented in a smart vest to constantly TABLE 1. Summary of healthcare IoT sensors in the literature.
monitor Chronic Obstructive Pulmonary Disease (COPD)
patients at home during the rest period between respiratory
rehabilitation exercises. This total solution system also pro-
vides an e-health platform based on the Internet of Medical
Things (IoMT) paradigm.
The body temperature is typically measured by thermistor-
type sensors to detect the conditions such as hypothermia,
heat stroke and fevers. In [17], a wearable, IoT cloud-based
system is presented for real-time personal health monitoring.
The collected data from the sensors can be viewed both on
a cloud dashboard as well as an embedded display on the
wearable. In [18], a tablet-shaped ingestible sensor is devel-
oped to measure the core-body temperature based on gastric
acid power generation. A custom integrated circuit (IC) is
prototyped which could wirelessly receive the temperature
data transferred by a tablet-shape device of diameter 10mm
and height 8mm.
Blood pressure (BP) is not considered a vital health sign
and is frequently measured alongside the other three vital
signs (i.e., pulse, respiratory and temperature). While most
BP wearable devices are non-invasive, they are still obstruc-
tive. In many cases a chest wearable ECG is connected to the
other sensors with some wires. As far as we are concerned
there is no comfortable wearable device developed to contin-
uously measure the BP with high accuracy. In [19], a survey
on commercial wearable IoT devices to monitor the BP from
a metrological point of view is presented. In this survey the
lack of the traceability and reliability of the BP measurements
is addressed.
Pulse oximeter sensors measure the blood oxygen using
PPG signals. The most widely used form of such sensors embedded into this platform to automatically evaluate speech
are the wrist-wearable pulse oximetry wearables. In [20], information without the need to preserve raw audio data.
a non-invasive wearable cardiac monitoring and alert sys- A summary of the aforementioned articles about healthcare
tem is proposed which can continuously measure the car- IoT sensors in given in Table 1.
diac values and using pulse oximetry. This solution detects
the saturation level of oxygen as well as blood volume B. ACTIVITY RECOGNITION AND SPORTS
variations in the tissues. The heart rate is obtained through This category is related to applications where the wearables
filtering and processing the sensed signals. The mobile com- are worn during sport activities to record different metrics
munication and GPS system enable emergency alerts when of the user/athlete activity in order to improve his/her per-
the measured cardiac values are out of the threshold val- formance Also, applications of this cluster consider gath-
ues. Moreover, the system gives this opportunity to the ering data regarding the recognition of daily activities of
user to self-trigger the alert system through Google voice humans and animals. Although activity recognition can have
assistant. some applications in medical diagnostics and out-of-hospital
Some other wearable sensors with applications in health- health-care, the applications belonging in this cluster cover
care could measure the blood glucose. For example, in [21], use cases beyond health category.
an ingestible smart drug is designed which circulates in the In this section, we discuss the use of wearable IoT for two
body and senses the glucose level by passing IR radiation. related activities. The first set of activities belong to recog-
The measured blood glucose level is then transmitted to the nition of daily physical activities [23] and the second one is
smartphone wirelessly. Another application of IoT wearable related to activities that are related to specific sports [24]. The
devices is in mental wellbeing monitoring where physiolog- recognition of daily physical activities is usually dedicated
ical signs are collected through behavioral traits. In [22], to tracking routines and body movement related to skeletal
an IoT-based wearable social sensing platform is developed muscles such as walking forward, walking backward, jog-
through integrating behavior monitoring, privacy audio fea- ging, sleeping, running, going up or down the stairs, bending
ture, and environment sensing in a naturalistic environment. waist, frontal elevation of arms, bending knees, and jumping
In particular, privacy protected audio-wellbeing features are front or back. It also includes recognition of static postures
such as sitting, sitting and relaxing, standing, laying down or tree [37]–[39], or artificial neural network and machine learn-
the recognition of going from one posture to another such as ing algorithm [40], [41].
going from standing posture to sitting one. For sport activities, measuring and estimating the perfor-
Generally speaking, IoT can bring an abundance of oppor- mance or efficiency of actions related to specific sport and
tunities for sport players, organizations and fans to increase providing feedback on parameters such as timing, angle or
their efficiency and offerings by creating an environment in amount of applying or releasing a force can help improve the
which athletes can receive better training or have access to accuracy and performance. Using these techniques, players
data that helps to keep them healthier, coaches are able to can get real-time feedback of their performance and improve
analyze injuries or find metrics on player performance, and their performance to make it more consistent. There are many
organizations can offer fan engagement strategies or allow papers that discuss the use of wearable devices for improving
fans to receive personalized offering from their favorite team. the quality of activities of a specific sport or even making the
Wearable IoT is a game changer in sport activities. However, wearer do something that would not be possible without wear-
the use of wearable IoT is currently limited due to some ing the wearable device. The wearer usually receives a mes-
leagues regulations and some challenges that need to be sage to help him/her to understand the quality of performance
resolved. Yet, as more and more organizations recognize the or provide help on how to proceed. These messages might be
advantages that IoT wearables can provide to players, coaches seen on the screen of the wearable or on displays in some
and fans, we will see that wearable IoT devices become other places. The wearer also may receive the messages or an
more and more prevalent in sports. Even though, wearable indication of a message using audio or haptics. The perfor-
IoT can provide endless opportunities to sports as explained mance of activities related to many different sports have been
above, in this section, we discuss those papers that are related researched such as badminton [42], [43], basketball [44],
to recognition of activities for specific sports such as ski- [45], rowing [46], [47], swimming [48], [49], hockey [50],
ing, climbing, martial arts, tennis, swimming, badminton, skiing [51], [52], martial art [53], [35], weightlifting [54],
weightlifting, or baseball. It should be mentioned that the tennis [55], baseball [56] and golf [57], [58].
other opportunities mentioned above have not received much
attention. C. TRACKING AND LOCALIZATION
The human physical activity detection has found lots of This category, is used mainly for tracking human and animal,
attention [25]–[30], [23]. To detect human physical activity to determine their location online. Finding the position of
one or more sensors are installed on one or several wearable a person or animal who is wearing a wearable device is
devices attached to the body. These sensors generate signals important in many applications. Studying the trajectory trip of
that can be analyzed to detect the type of activity and find the a bird, finding the location of a senior citizen in a care-home
information that differentiates various activities. The existing facility, analyzing the movement of the people who are vis-
methods usually achieve this goal in four stages. The first iting an exhibition, or pet tracking are some examples of
stage is pre-processing, noise cancelation and signal range these applications. A comprehensive study on localization
adjustment which prepares the signal for the next stages. using IoT technology is published in [59], but the study does
The second stage is feature extraction in which the fea- not concentrate on wearable IoT. In general, the localization
tures or parameters specific to the signal are extracted. The methods discussed in literature can be divided into two main
extracted features can usually be categorized as structural categories. The ones that use offline training and the ones
features or statistical ones. The structural features find the that without using offline training can determine the location.
correlation among different signals. The statistical features The training dependent methods can be classified into three
can be found through time- or transformed-domain analysis clusters: 1) Fingerprinting 2) Stochastic-oriented models and
of a signal. Example of statistical features are mean value, 3) Machine learning based schemes. Fingerprinting has been
max value, peak, and size of signal. The transformation studied extensively using various types of signals includ-
typically used for finding the statistical features are Fourier ing general signal pattern, audio signal, video signal and
Transform (FT) [31], [32], Discrete Cosine Transform (DCT) motion [60], [61]. General signal pattern usually depends
[33], or Wavelet Transform (WT). The third stage is feature on the Received Signal Strength Indicator (RSSI) value of
reduction which tries to find most important features that the wireless signal in which the location fingerprint is found
can be used in final stage by removing the redundant and through site survey and is recorded in a fingerprint data
irrelevant information. Classification is the last stage in which base. This fingerprint information is used later by a localiza-
the reduced set of features are clustered to find patterns tion algorithm online to estimate the location. Fingerprinting
among various activities using a classification algorithm. can be used on visual information captured by a camera or
Classification algorithms can be done using the probability audio signals captured by a microphone. The IoT devices can
of an activity according to a set of features and their prob- also use collaborative localization methods that are based on
abilities [34], the simulates between features according to a mutual position measurement of each IoT device as well as
dataset, transformation of features to a space that can help the relative distance of devices.
in identifying the type of activity [35], [36], hierarchy based The non-training dependent category of localization
model to map the features to activities based on a decision method can also be divided into various clusters. The main
3) WEARABILITY
The wearable IoT devices have to be comfortable when worn
by the user. It is important that they are light and designed
in such a way that they do not disturb the normal activi-
ties of the user. The tradeoff between the complexity of the
FIGURE 5. Major challenges of wearable IoT technology.
computations and the weight of the wearable is one of the
major challenges. In [109], the idea of smart clothing or
wearable 2.0 for human-cloud integration is introduced which
to use their cellphone even in places where landline systems
tries to solve the problem with discomfort caused by wearing
were available. We have seen the same user behavior in data
multiple sensors in different part of the body for health care
usage over the Internet. In the past, due to the cost of data
application.
usage and communication speed, users preferred to switch to
Wi-Fi and connect to hotspots instead of cellular connection. 4) SAFETY
However, as the cost of data usage is decreasing, more and
All wearable IoT devices use wireless technologies to trans-
more mobile users use their cellular connection to connect to
mit their sensed data to another node, gateway or a base
the Internet in places where Wi-Fi connection is also avail-
station. This wireless transmission involves radiofrequency
able. Therefore, while there exist some IoT enabling tech-
radiation which could have negative impact on the user’s
nologies today that may be used for wearable devices in local
health because the transceiver antennas are very close to
area or solve the wide area coverage requirement of the IoT
his/her body. In those wearables that are worn on the head
devices by using a cellular connection through a cellphone,
or eyes the radiation risks could be significantly higher. This
they fall short as compared to two existing CIoT technologies
safety concern is well addressed in [110] by reviewing the
of the 3rd Generation Partnership Project (3GPP), which are
standard limits of human exposure to radio frequency elec-
LTE-M and Narrow Band IoT (NB-IoT) in terms of coverage,
tromagnetic energy and analyzing the radiation level of CIoT
scalability, interoperability, QoS, and security. LTE-M and
antennas. It is shown that the problem can be worse when
NB-IoT are introduced in Release 13 of LTE technology.
the wearable CIoT device is transmitting in areas with poor
In LTE Release, 14, and 15, the enhancements of LTE IoT
coverage.
continued to provide cellular IoT connectivity to more IoT
devices and in more diverse applications. It should be noted 5) SECURITY
that the performance of CIoT will further hugely be advanced
The complexity of the wearable IoT devices is typically
with more implementation of 5G cellular technology.
reduced due to lightweight and less power consuming
designs. Consequently, there could be less strong security fea-
IV. IOT- ENABLED WEARABLE CHALLENGES AND tures on such devices. One of the challenges in wearable IoT
FUTURE POSSIBILITIES devices is how to implement security policies while keeping
The major challenges of the wearable IoT devices as shown the complexity of the system as low as possible. In general,
in Figure 5, are listed below: wearables are easy hacking targets due to poor encryption and
protection.
1) DATA RESOLUTION OF WEARABLE SENSORS
Since it is of high importance that the wearable device must 6) REGULATION
be comfortable when worn by the user and should consume There is currently a limitation in using wearable IoT devices
low amount of energy, they are typically small in size and in many industries due to lack of existence of proper regu-
the sensors have lower resolution compared to non-wearable lations. For example, in sports fields and arenas, the use of
devices. wearable IoT devices are technologically feasible, but it is
not being used due to the leagues regulations.
2) POWER CONSUMPTION
To minimize the human interaction and for wearable devices 7) PRIVACY
to operate for long hours without replacing or charging The constant exchange of personal data such as vital health
the battery, special considerations are to be taken into signals, dosage, and location between the wearable and the
account while designing the wearables. For example, low IoT hub can create an environment for privacy breaches.
power consumption systems or energy harvesting techniques Typically, wearable IoT devices are on broadcast mode which
makes them easily discoverable by other nodes in the net- [6] A. Ghorbel, S. Bouguerra, N. B. Amor, and M. Jallouli, ‘‘Cloud based
work. Unauthorized nodes can steal the personal data if mobile application for remote control of intelligent wheelchair,’’ in Proc.
14th Int. Wireless Commun. Mobile Comput. Conf. (IWCMC), Limassol,
appropriate privacy policies are not applied. In such broadcast Cyprus, Jun. 2018, pp. 1249–1254.
modes, the built-in hardware security technology of the IoT [7] M. O. Agyeman and A. Al-Mahmood, ‘‘Design and implementation of
devices might not guarantee the protection of personal data a wearable device for motivating patients with upper and/or lower limb
disability via gaming and home rehabilitation,’’ in Proc. 4th Int. Conf. Fog
against breaches. In [111] a broadcast-subscriber IoT model Mobile Edge Comput. (FMEC), Jun. 2019, pp. 247–252.
is proposed where the users’ personal data are only shared [8] C. F. Pasluosta, H. Gassner, J. Winkler, J. Klucken, and B. M. Eskofier,
with intended nodes such as healthcare facilities or devices ‘‘An emerging era in the management of Parkinson’s disease: Wearable
technologies and the Internet of Things,’’ IEEE J. Biomed. Health Infor-
authorized by the user. mat., vol. 19, no. 6, pp. 1873–1881, Nov. 2015.
There will be enormous number of applications and oppor- [9] S. B. Baker, W. Xiang, and I. Atkinson, ‘‘Internet of Things for smart
tunities for wearable IoT devices. Using 3D printing, the next healthcare: Technologies, challenges, and opportunities,’’ IEEE Access,
vol. 5, pp. 26521–26544, 2017.
generation of these wearables will be boosted by lateraliza- [10] S. Jayanth, M. B. Poorvi, R. Shreyas, B. Padmaja, and M. P. Sunil,
tion of their design, fabrication, and distribution. For example ‘‘Wearable device to measure heart beat using IoT,’’ in Proc. Int. Conf.
by using this technology for health application, the patient Inventive Syst. Control (ICISC), Coimbatore, India, Jan. 2017, pp. 1–5.
[11] A. J. A. Majumder, Y. A. ElSaadany, R. Young, and D. R. Ucci,
can upload their disease profiles and download his/her indi- ‘‘An energy efficient wearable smart IoT system to predict cardiac arrest,’’
vidualized treatment tools to print them at home with low Adv. Hum.-Comput. Interact., vol. 2019, pp. 1–21, Feb. 2019.
cost. By improvement in new energy harvesting technologies [12] A. Brezulianu, O. Geman, M. D. Zbancioc, M. Hagan, C. Aghion,
D. J. Hemanth, and L. H. Son, ‘‘IoT based heart activity monitoring using
and existence of longer life batteries, the greater computa- inductive sensors,’’ Sensors, vol. 19, no. 15, p. 3284, 2019.
tional power will be allowed in wearable IoT devices. This [13] S. Milici, J. Lorenzo, A. Lazaro, R. Villarino, and D. Girbau, ‘‘Wireless
increases their autonomy, implementation of feature extrac- breathing sensor based on wearable modulated frequency selective sur-
face,’’ IEEE Sensors J., vol. 17, no. 5, pp. 1285–1292, Mar. 2017.
tion and classification and/or prediction, and allows an effi- [14] S. Syed Tauhid, B. Faizan, D. Faheem, A. Nouman, and A. Jan, ‘‘Cloud-
cient employment of hardware resources. The most important assisted IoT-based smart respiratory monitoring system for asthma
opportunity in werable IoT will be realized when IoT data patients,’’ in Applications of Intelligent Technologies in Healthcare.
New York, NY, USA: Springer, 2019.
from various sources and devices are combined there will be [15] I. Mahbub, S. A. Pullano, H. Wang, S. K. Islam, A. S. Fiorillo, G. To,
an integrated IoT system available. and M. R. Mahfouz, ‘‘A low-power wireless piezoelectric sensor-based
respiration monitoring system realized in CMOS process,’’ IEEE Sensors
J., vol. 17, no. 6, pp. 1858–1864, Mar. 2017.
V. CONCLUSION
[16] D. Naranjo-Hernández, A. Talaminos-Barroso, J. Reina-Tosina, L. Roa,
Wearable IoT can offer endless new opportunities in many G. Barbarov-Rostan, P. Cejudo-Ramos, E. Márquez-Martín, and
real-life applications. This paper provided a survey on the F. Ortega-Ruiz, ‘‘Smart vest for respiratory rate monitoring of COPD
patients based on non-contact capacitive sensing,’’ Sensors, vol. 18,
most important efforts of the research community in the area no. 7, p. 2144, 2018.
of wearable and IoT. After identifying more than one hundred [17] J. Wan, M. A. A. H. Al-awlaqi, M. Li, M. O’Grady, X. Gu, J. Wang,
papers as the significant literature in this area, the papers were and N. Cao, ‘‘Wearable IoT enabled real-time health monitoring system,’’
EURASIP J. Wireless Commun. Netw., vol. 2018, no. 1, Dec. 2018.
analyzed and divided in four major clusters based on their [18] S. Yoshida, H. Miyaguchi, and T. Nakamura, ‘‘Development of
applications. The methods used for each cluster also were tablet-shaped ingestible core-body thermometer powered by gastric
grouped. acid battery,’’ IEEE Sensors J., vol. 18, no. 23, pp. 9755–9762,
Dec. 2018.
Wearables have lots of potential when integrated IoT sys- [19] F. Lamonaca, E. Balestrieri, I. Tudosa, F. Picariello, D. L. Carni,
tem becomes available. For this reason the true power of C. Scuro, F. Bonavolonta, V. Spagnuolo, G. Grimaldi, and A. Colaprico,
combining wearable and IoT has not been recognized. Also ‘‘An overview on Internet of medical things in blood pressure monitor-
ing,’’ in Proc. IEEE Int. Symp. Med. Meas. Appl. (MeMeA), Istanbul,
cellular IoT can revolutionize the wearable IoT industry Turkey, Jun. 2019, pp. 1–6.
which at this point has not received lots of attention among [20] D. Murali, D. R. Rao, S. R. Rao, and M. Ananda, ‘‘Pulse oximetry
research community. and IOT based cardiac monitoring integrated alert system,’’ in Proc. Int.
Conf. Adv. Comput., Commun. Informat. (ICACCI), Bangalore, India,
Sep. 2018, pp. 2237–2243.
REFERENCES [21] B. Sargunam and S. Anusha, ‘‘IoT based mobile medical applica-
[1] S. Seneviratne, Y. Hu, T. Nguyen, G. Lan, S. Khalifa, K. Thilakarathna, tion for smart insulin regulation,’’ in Proc. IEEE Int. Conf. Electr.,
M. Hassan, and A. Seneviratne, ‘‘A survey of wearable devices and Comput. Commun. Technol. (ICECCT), Coimbatore, India, Feb. 2019,
challenges,’’ IEEE Commun. Surveys Tuts., vol. 19, no. 4, pp. 2573–2620, pp. 1–5.
4th Quart., 2017. [22] S. Yang, B. Gao, L. Jiang, J. Jin, Z. Gao, X. Ma, and W. L. Woo, ‘‘IoT
[2] K. Hartman, Make: Wearable Electronics: Design, Prototype, and Wear structured long-term wearable social sensing for mental wellbeing,’’ IEEE
Your Own Interactive Garments. Sebastopol, CA, USA: Maker Media, Internet Things J., vol. 6, no. 2, pp. 3652–3662, Apr. 2019.
2014. [23] J. Qi, P. Yang, M. Hanneghan, S. Tang, and B. Zhou, ‘‘A hybrid hier-
[3] E. Sazonov and M. R. Neuman, Wearable Sensors: Fundamentals, Imple- archical framework for gym physical activity recognition and measure-
mentation and Applications. Amsterdam, The Netherlands: Elsevier, ment using wearable sensors,’’ IEEE Internet Things J., vol. 6, no. 2,
2014. pp. 1384–1393, Apr. 2019.
[4] C. Nave and O. Postolache, ‘‘Smart walker based IoT physical rehabilita- [24] E. Mencarini, A. Rapp, L. Tirabeni, and M. Zancanaro, ‘‘Designing
tion system,’’ in Proc. Int. Symp. Sens. Instrum. IoT Era (ISSI), Shanghai, wearable systems for sports: A review of trends and opportunities in
China, Sep. 2018, pp. 1–6. human–computer interaction,’’ IEEE Trans. Hum.-Mach. Syst., vol. 49,
[5] G. Yang, J. Deng, G. Pang, H. Zhang, J. Li, B. Deng, Z. Pang, J. Xu, no. 4, pp. 314–325, Aug. 2019.
M. Jiang, P. Liljeberg, H. Xie, and H. Yang, ‘‘An IoT-enabled stroke [25] D. Castro, W. Coral, C. Rodriguez, J. Cabra, and J. Colorado, ‘‘Wearable-
rehabilitation system based on smart wearable armband and machine based human activity recognition using an IoT approach,’’ J. Sensor
learning,’’ IEEE J. Transl. Eng. Health Med., vol. 6, pp. 1–10, 2018. Actuator Netw., vol. 6, no. 4, p. 28, 2017.
[26] H. Huang, X. Li, S. Liu, S. Hu, and Y. Sun, ‘‘TriboMotion: A self- [47] Z. Wang, J. Wang, H. Zhao, N. Yang, and G. Fortino, ‘‘CanoeSense:
powered triboelectric motion sensor in wearable Internet of Things for Monitoring canoe sprint motion using wearable sensors,’’ in Proc. IEEE
human activity recognition and energy harvesting,’’ IEEE Internet Things Int. Conf. Syst., Man, Cybern. (SMC), Budapest, Hungary, Oct. 2016,
J., vol. 5, no. 6, pp. 4441–4453, Dec. 2018. pp. 644–649.
[27] W. Lu, F. Fan, J. Chu, P. Jing, and S. Yuting, ‘‘Wearable computing [48] J. Pansiot, B. Lo, and G.-Z. Yang, ‘‘Swimming stroke kinematic analysis
for Internet of Things: A discriminant approach for human activity with BSN,’’ in Proc. Int. Conf. Body Sensor Netw., Singapore, Jun. 2010,
recognition,’’ IEEE Internet Things J., vol. 6, no. 2, pp. 2749–2759, pp. 153–158.
Apr. 2019. [49] M. Bächlin, K. Förster, and G. Tröster, ‘‘SwimMaster: A wearable
[28] L. Atallah, B. Lo, R. King, and G.-Z. Yang, ‘‘Sensor positioning for activ- assistant for swimmer,’’ in Proc. 11th Int. Conf. Ubiquitous Comput.,
ity recognition using wearable accelerometers,’’ IEEE Trans. Biomed. Sep. 2009, pp. 215–224.
Circuits Syst., vol. 5, no. 4, pp. 320–329, Aug. 2011. [50] J. Häkkilä, M. Alhonsuo, L. Virtanen, J. Rantakari, A. Colley, and
[29] M. J. Mathie, B. G. Celler, N. H. Lovell, and A. C. F. Coster, ‘‘Classifica- T. Koivumäki, ‘‘MyData approach for personal health—A service design
tion of basic daily movements using a triaxial accelerometer,’’ Med. Biol. case for young athletes,’’ in Proc. 49th Hawaii Int. Conf. Syst. Sci.
Eng. Comput., vol. 42, no. 5, pp. 679–687, Sep. 2004. (HICSS), Koloa, HI, USA, Jan. 2016, pp. 3493–3502.
[30] A. Mannini and A. M. Sabatini, ‘‘Machine learning methods for classi- [51] J. Haladjian, M. Reif, and B. Brügge, ‘‘VIHapp: A wearable system to
fying human physical activity from on-body accelerometers,’’ Sensors, support blind skiing,’’ in Proc. ACM Int. Joint Conf. Pervas. Ubiquitous
vol. 10, no. 2, pp. 1154–1175, 2010. Comput. (UbiComp), 2017, pp. 1033–1037.
[31] K. H. Walse, R. V. Dharaskar, and V. M. Thakare, ‘‘A study on the effect [52] E. Niforatos, A. Fedosov, I. Elhart, and M. Langheinrich, ‘‘Augmenting
of adaptive boosting on performance of classifiers for human activity skiers’ peripheral perception,’’ in Proc. ACM Int. Symp. Wearable Com-
recognition,’’ in Proc. Int. Conf. Data Eng. Commun. Technol., 2017, put. (ISWC), 2017, pp. 114–121.
pp. 419–429. [53] E. H. Chi, J. Song, and G. Corbin, ‘‘‘Killer App’ of wearable comput-
[32] Y. Chen and C. Shen, ‘‘Performance analysis of smartphone-sensor ing: Wireless force sensing body protectors for martial arts,’’ in Proc.
behavior for human activity recognition,’’ IEEE Access, vol. 5, 17th Annu. ACM Symp. User Interface Softw. Technol. (UIST), 2004,
pp. 3095–3110, 2017. pp. 277–285.
[33] Z. He and L. Jin, ‘‘Activity recognition from acceleration data based on [54] M. Pan, S. Salvi, and E. Brady, ‘‘Designing auditory feedback from
discrete consine transform and SVM,’’ in Proc. IEEE Int. Conf. Syst., wearable weightlifting devices,’’ in Proc. Extended Abstr. CHI Conf.
Man, Cybern., Oct. 2009, pp. 5041–5044. Hum. Factors Comput. Syst. (CHI), 2018, pp. 1–6.
[34] L. Liu, S. Wang, G. Su, Z.-G. Huang, and M. Liu, ‘‘Towards com- [55] H. Havlucu, I. Bostan, A. Coskun, and Ö. Özcan, ‘‘Understanding the
plex activity recognition using a Bayesian network-based probabilis- lonesome tennis players: Insights for future wearables,’’ in Proc. CHI
tic generative framework,’’ Pattern Recognit., vol. 68, pp. 295–309, Conf. Extended Abstr. Hum. Factors Comput. Syst. (CHI EA), 2017,
Aug. 2017. pp. 1678–1685.
[56] M. Lapinski, E. Berkson, T. Gill, M. Reinold, and J. A. Paradiso, ‘‘A dis-
[35] K.-C. Liu, C.-Y. Yen, L.-H. Chang, C.-Y. Hsieh, and C.-T. Chan, ‘‘Wear-
tributed wearable, wireless sensor system for evaluating professional
able sensor-based activity recognition for housekeeping task,’’ in Proc.
baseball pitchers and batters,’’ in Proc. Int. Symp. Wearable Comput.,
IEEE 14th Int. Conf. Wearable Implant. Body Sensor Netw. (BSN),
Sep. 2009, pp. 131–138.
May 2017, pp. 67–70.
[57] D. K. Arvind and A. Bates, ‘‘The speckled golfer,’’ in Proc. 3rd Int. ICST
[36] B. Logan, J. Healey, M. Philipose, E. M. Tapia, and S. Intille, ‘‘A long-
Conf. Body Area Netw., 2008, pp. 1–28
term evaluation of sensing modalities for activity recognition,’’ in Proc.
[58] H. Ghasemzadeh, V. Loseu, E. Guenterberg, and R. Jafari, ‘‘Sport training
UbiComp Ubiquitous Comput., Innsbruck, Austria, vol. 4717, 2007,
using body sensor networks: A statistical approach to measure wrist
pp. 483–500.
rotation for golf swing,’’ in Proc. 4th Int. ICST Conf. Body Area Netw.,
[37] S.-M. Lee, S. Min Yoon, and H. Cho, ‘‘Human activity recognition from
Los Angeles, CA, USA, Apr. 2009, pp. 1–8.
accelerometer data using convolutional neural network,’’ in Proc. IEEE
[59] R. C. Shit, S. Sharma, D. Puthal, and A. Y. Zomaya, ‘‘Location of things
Int. Conf. Big Data Smart Comput. (BigComp), Feb. 2017, pp. 131–134.
(LoT): A review and taxonomy of sensors localization in IoT infras-
[38] M. Ermes, J. Pärkkä, J. Mäntyjärvi, and I. Korhonen, ‘‘Detection of daily tructure,’’ IEEE Commun. Surveys Tuts., vol. 20, no. 3, pp. 2028–2061,
activities and sports with wearable sensors in controlled and uncontrolled 3rd Quart., 2018.
conditions,’’ IEEE Trans. Inf. Technol. Biomed., vol. 12, no. 1, pp. 20–26, [60] Y. Kim, H. Shin, and H. Cha, ‘‘Smartphone-based Wi-Fi pedestrian-
Jan. 2008. tracking system tolerating the RSS variance problem,’’ in Proc. IEEE
[39] M. Li, V. Rozgic, G. Thatte, S. Lee, A. Emken, M. Annavaram, U. Mitra, Int. Conf. Pervas. Comput. Commun., Lugano, Switzerland, Mar. 2012,
D. Spruijt-Metz, and S. Narayanan, ‘‘Multimodal physical activity recog- pp. 11–19.
nition by fusing temporal and cepstral information,’’ IEEE Trans. Neural [61] A. Hallquist and A. Zakhor, ‘‘Single view pose estimation of mobile
Syst. Rehabil. Eng., vol. 18, no. 4, pp. 369–380, Aug. 2010. devices in urban environments,’’ in Proc. IEEE Workshop Appl. Comput.
[40] S. Liu, R. X. Gao, D. John, J. W. Staudenmayer, and P. S. Freedson, Vis. (WACV), Tampa, FL, USA, Jan. 2013, pp. 347–354.
‘‘Multisensor data fusion for physical activity assessment,’’ IEEE Trans. [62] B. Wang, G. Wu, S. Wang, and L. T. Yang, ‘‘Localization based on
Biomed. Eng., vol. 59, no. 3, pp. 687–696, Mar. 2012. adaptive regulated neighborhood distance for wireless sensor networks
[41] I. C. Gyllensten and A. G. Bonomi, ‘‘Identifying types of physical activity with a general radio propagation model,’’ IEEE Sensors J., vol. 14, no. 11,
with a single accelerometer: Evaluating laboratory-trained algorithms in pp. 3754–3762, Nov. 2014.
daily life,’’ IEEE Trans. Biomed. Eng., vol. 58, no. 9, pp. 2656–2663, [63] H. Chen, P. Huang, M. Martins, H. C. So, and K. Sezaki, ‘‘Novel cen-
Sep. 2011. troid localization algorithm for three-dimensional wireless sensor net-
[42] Z. Wang, M. Guo, and C. Zhao, ‘‘Badminton stroke recognition based on works,’’ in Proc. 4th Int. Conf. Wireless Commun., Netw. Mobile Comput.,
body sensor networks,’’ IEEE Trans. Hum.-Mach. Syst., vol. 46, no. 5, Oct. 2008, pp. 1–4.
pp. 769–775, Oct. 2016. [64] H. Yu, X. Chen, and J. Fan, ‘‘Gauss-newton method based on energy
[43] A. Raina, T. G. Lakshmi, and S. Murthy, ‘‘CoMBaT: Wearable technology target localization,’’ Comput. Eng. Appl., vol. 43, no. 27, pp. 124–126,
based training system for novice badminton players,’’ in Proc. IEEE 17th 2007.
Int. Conf. Adv. Learn. Technol. (ICALT), Jul. 2017, pp. 153–157. [65] L. Lazos and R. Poovendran, ‘‘HiRLoc: High-resolution robust localiza-
[44] J. C. Maglott, J. Xu, and P. B. Shull, ‘‘Differences in arm motion timing tion for wireless sensor networks,’’ IEEE J. Sel. Areas Commun., vol. 24,
characteristics for basketball free throw and jump shooting via a body- no. 2, pp. 233–246, Feb. 2006.
worn sensorized sleeve,’’ in Proc. IEEE 14th Int. Conf. Wearable Implant. [66] W. Jichun, H. Liusheng, X. Hongli, X. Ben, and L. Shanliang, ‘‘A novel
Body Sensor Netw. (BSN), May 2017, pp. 31–34. range free localization scheme based on Voronoi diagrams in wireless
[45] S. Bogers, C. Megens, and S. Vos, ‘‘Design for balanced engagement sensor networks,’’ J. Comput. Res. Develop., vol. 5, no. 1, pp. 119–125,
in mixed level sports teams,’’ in Proc. CHI Conf. Extended Abstr. Hum. 2008.
Factors Comput. Syst. (CHI EA), 2017, pp. 994–1002. [67] B. Mirmahboub, S. Samavi, N. Karimi, and S. Shirani, ‘‘Automatic
[46] Z. Wang, H. Zhao, S. Qiu, and Q. Gao, ‘‘Stance-phase detection for monocular system for human fall detection based on variations in sil-
ZUPT-aided foot-mounted pedestrian navigation system,’’ IEEE/ASME houette area,’’ IEEE Trans. Biomed. Eng., vol. 60, no. 2, pp. 427–436,
Trans. Mechatronics, vol. 20, no. 6, pp. 3170–3181, Dec. 2015. Feb. 2013.
[68] E. Auvinet, F. Multon, A. Saint-Arnaud, J. Rousseau, and J. Meunier, [90] D. Antolín, N. Medrano, B. Calvo, and F. Pérez, ‘‘A wearable wireless
‘‘Fall detection with multiple cameras: An occlusion-resistant method sensor network for indoor smart environment monitoring in safety appli-
based on 3-D silhouette vertical distribution,’’ IEEE Trans. Inf. Technol. cations,’’ Sensors, vol. 17, no. 2, p. 365, 2017.
Biomed., vol. 15, no. 2, pp. 290–300, Mar. 2011. [91] E. Wilhelm, S. Siby, Y. Zhou, X. J. S. Ashok, M. Jayasuriya, S. Foong,
[69] C. Rougier, J. Meunier, A. St-Arnaud, and J. Rousseau, ‘‘Robust video J. Kee, K. L. Wood, and N. O. Tippenhauer, ‘‘Wearable environmental
surveillance for fall detection based on human shape deformation,’’ sensors and infrastructure for mobile large-scale urban deployment,’’
IEEE Trans. Circuits Syst. Video Technol., vol. 21, no. 5, pp. 611–622, IEEE Sensors J., vol. 16, no. 22, pp. 8111–8123, Nov. 2016.
May 2011. [92] M. Serbanescu, V. M. Placinta, O. E. Hutanu, and C. Ravariu, ‘‘Smart, low
[70] M. Daher, A. Diab, M. El Badaoui El Najjar, M. A. Khalil, and power, wearable multi-sensor data acquisition system for environmental
F. Charpillet, ‘‘Elder tracking and fall detection system using smart tiles,’’ monitoring,’’ in Proc. 10th Int. Symp. Adv. Topics Electr. Eng. (ATEE),
IEEE Sensors J., vol. 17, no. 2, pp. 469–479, Jan. 2017. Bucharest, Romania, 2017, pp. 118–123.
[71] W. Saadeh, S. A. Butt, and M. A. B. Altaf, ‘‘A patient-specific sin- [93] J. R. Byrne, K. O’Sullivan, and K. Sullivan, ‘‘An IoT and wearable
gle sensor IoT-based wearable fall prediction and detection system,’’ technology hackathon for promoting careers in computer science,’’ IEEE
IEEE Trans. Neural Syst. Rehabil. Eng., vol. 27, no. 5, pp. 995–1003, Trans. Educ., vol. 60, no. 1, pp. 50–58, Feb. 2017.
May 2019. [94] E. de la Guía, V. L. Camacho, L. Orozco-Barbosa, V. M. B. Luján,
[72] A. Sorvala, E. Alasaarela, H. Sorvoja, and R. Myllylä, ‘‘A two-threshold V. M. R. Penichet, and M. L. Peréz, ‘‘Introducing IoT and wearable tech-
fall detection algorithm for reducing false alarms,’’ in Proc. 6th Int. Symp. nologies into task-based language learning for young children,’’ IEEE
Med. Inf. Commun. Technol. (ISMICT), Mar. 2012, pp. 1–4. Trans. Learn. Technol., vol. 9, no. 4, pp. 366–378, Oct. 2016.
[73] W. Saadeh, M. A. Bin Altaf, and S. A. Butt, ‘‘A wearable neuro- [95] A. G. Perez, D. Lobo, F. Chinello, G. Cirio, M. Malvezzi, J. S. Martin,
degenerative diseases detection system based on gait dynamics,’’ in Proc. D. Prattichizzo, and M. A. Otaduy, ‘‘Optimization-based wearable tactile
IFIP/IEEE Int. Conf. Very Large Scale Integr. (VLSI-SoC), Oct. 2017, rendering,’’ IEEE Trans. Haptics, vol. 10, no. 2, pp. 254–264, Apr. 2017.
pp. 1–6. [96] M. Maisto, C. Pacchierotti, F. Chinello, G. Salvietti, A. De Luca, and
[74] R. Bartlett, Introduction to Sports Biomechanics: Analysing Human D. Prattichizzo, ‘‘Evaluation of wearable haptic systems for the fingers
Movement Patterns, 2nd ed. Abingdon, U.K: Routledge, 2007. in augmented reality applications,’’ IEEE Trans. Haptics, vol. 10, no. 4,
[75] L. Ren and Y. Peng, ‘‘Research of fall detection and fall prevention tech- pp. 511–522, Oct. 2017.
nologies: A systematic review,’’ IEEE Access, vol. 7, pp. 77702–77722, [97] R. Silberglitt, A. Lauland, M. Watson, C. A. Eusebi, and J. Lastunen.
2019. Weaable Technology and Law Enforcement. Accessed: Dec. 10, 2019.
[76] N. Otanasap, ‘‘Pre-impact fall detection based on wearable device [Online]. Available: https://www.rand.org/content/dam/rand/
using dynamic threshold model,’’ in Proc. 17th Int. Conf. Parallel Dis- pubs/research_reports/RR2000/RR2012/RAND_RR2012.pdf
trib. Comput., Appl. Technol. (PDCAT), Guangzhou, China, Dec. 2016,
[98] C. Wang, J. Liu, X. Guo, Y. Wang, and Y. Chen, ‘‘WristSpy: Snooping
pp. 362–365.
passcodes in mobile payment using wrist-worn wearables,’’ in Proc. IEEE
[77] Y. Wu, Y. Su, Y. Hu, N. Yu, and R. Feng, ‘‘A multi-sensor fall detection Conf. Comput. Commun. (INFOCOM), Apr. 2019, pp. 2071–2079.
system based on multivariate statistical process analysis,’’ J. Med. Biol.
[99] F. J. Dian, A. Yousefi, and S. Lim, ‘‘A practical study on Bluetooth
Eng., vol. 39, no. 3, pp. 336–351, Jun. 2019.
low energy (BLE) throughput,’’ in Proc. IEEE 9th Annu. Inf. Technol.,
[78] L. Ren and W. Shi, ‘‘Chameleon: Personalised and adaptive fall detection
Electron. Mobile Commun. Conf. (IEMCON), Vancouver, BC, Canada,
of elderly people in home-based environments,’’ Int. J. Sensor Netw.,
Nov. 2018, pp. 768–771.
vol. 20, no. 3, pp. 163–176, 2016.
[100] F. J. Dian, ‘‘Low-power synchronized multi-channel data acquisition
[79] M. N. Nyan, F. E. H. Tay, and E. Murugasu, ‘‘A wearable system for pre-
communication system,’’ in Proc. IEEE 9th Annu. Comput. Com-
impact fall detection,’’ J. Biomechanics, vol. 41, no. 16, pp. 3475–3481,
mun. Workshop Conf. (CCWC), Las Vegas, NV, USA, Jan. 2019,
Dec. 2008.
pp. 1027–1031.
[80] S. Yu, H. Chen, and R. A. Brown, ‘‘Hidden Markov model-based fall
detection with motion sensor orientation calibration: A case for real-life [101] F. J. Dian, A. Yousefi, and K. Somaratne, ‘‘A study in accuracy of
home monitoring,’’ IEEE J. Biomed. Health Informat., vol. 22, no. 6, time synchronization of BLE devices using connection-based event,’’ in
pp. 1847–1853, Nov. 2018. Proc. 8th IEEE Annu. Inf. Technol., Electron. Mobile Commun. Conf.
(IEMCON), Vancouver, BC, Canada, Oct. 2017, pp. 595–601.
[81] X. Zhou, L.-C. Qian, P.-J. You, Z.-G. Ding, and Y.-Q. Han, ‘‘Fall detection
using convolutional neural network with multi-sensor fusion,’’ in Proc. [102] A. Yousefi, K. Somaratne, and F. J. Dian, ‘‘Analysis of time synchro-
IEEE Int. Conf. Multimedia Expo Workshops (ICMEW), San Diego, CA, nization based on current measurement for Bluetooth low energy (BLE),’’
USA, Jul. 2018, pp. 1–5. in Proc. 8th IEEE Annu. Inf. Technol., Electron. Mobile Commun. Conf.
[82] W. Min, H. Cui, H. Rao, Z. Li, and L. Yao, ‘‘Detection of human falls (IEMCON), Vancouver, BC, Canada, Oct. 2017, pp. 602–607.
on furniture using scene analysis based on deep learning and activity [103] K. Somaratne, F. J. Dian, and A. Yousefi, ‘‘Accuracy analysis of time
characteristics,’’ IEEE Access, vol. 6, pp. 9324–9335, 2018. synchronization using current consumption pattern of BLE devices,’’
[83] J. K. Lee, S. N. Robinovitch, and E. J. Park, ‘‘Inertial sensing-based in Proc. IEEE 8th Annu. Comput. Commun. Workshop Conf. (CCWC),
pre-impact detection of falls involving near-fall scenarios,’’ IEEE Trans. Las Vegas, NV, USA, Jan. 2018, pp. 841–844.
Neural Syst. Rehabil. Eng., vol. 23, no. 2, pp. 258–266, Mar. 2015. [104] F. J. Dian, A. Yousefi, and K. Somaratne, ‘‘Performance evaluation of
[84] G. Shi, C. S. Chan, W. J. Li, K.-S. Leung, Y. Zou, and Y. Jin, ‘‘Mobile time synchronization using current consumption pattern of BLE devices,’’
human airbag system for fall protection using MEMS sensors and in Proc. IEEE 8th Annu. Comput. Commun. Workshop Conf. (CCWC),
embedded SVM classifier,’’ IEEE Sensors J., vol. 9, no. 5, pp. 495–503, Las Vegas, NV, USA, Jan. 2018, pp. 906–910.
May 2009. [105] F. J. Dian, ‘‘An analytical scheme for power consumption of battery-
[85] S. Shan and T. Yuan, ‘‘A wearable pre-impact fall detector using feature operated peripheral BLE nodes,’’ in Proc. IEEE 9th Annu. Comput.
selection and support vector machine,’’ in Proc. IEEE 10th Int. Conf. Commun. Workshop Conf. (CCWC), Las Vegas, NV, USA, Jan. 2019,
SIGNAL Process. Proc., Beijing, China, Oct. 2010, pp. 1686–1689. pp. 1021–1026.
[86] P. Choudhary, R. Sharma, G. Singh, and S. Das, ‘‘A survey paper on [106] F. J. Dian, A. Yousefi, and S. Lim, ‘‘Time scheduling of central BLE
drowsiness detection & alarm system for drivers,’’ Int. Res. J. Eng. for connection events,’’ in Proc. IEEE 9th Annu. Inf. Technol., Electron.
Technol. (IRJET), vol. 3, no. 12, pp. 1433–1437, 2016. Mobile Commun. Conf. (IEMCON), Vancouver, CO, USA, Nov. 2018,
[87] W.-J. Chang, L.-B. Chen, and Y.-Z. Chiou, ‘‘Design and implementa- pp. 763–767.
tion of a Drowsiness-Fatigue-Detection system based on wearable smart [107] F. J. Dian and R. Vahidnia, ‘‘Formulation of BLE throughput based
glasses to increase road safety,’’ IEEE Trans. Consum. Electron., vol. 64, on node and link parameters,’’ IEEE Can. J. Elect. Comput. Eng.,
no. 4, pp. 461–469, Nov. 2018. 2020.
[88] S. Dhole, ‘‘A novel helmet design and implementation for drowsiness and [108] T. Wu, F. Wu, J.-M. Redoute, and M. R. Yuce, ‘‘An autonomous wireless
fall detection of workers on-site using EEG and random-forest classifier,’’ body area network implementation towards IoT connected healthcare
in Proc. 10th Int. Conf. Ambient Syst., Netw. Technol., 2019, pp. 947–952. applications,’’ IEEE Access, vol. 5, pp. 11413–11422, 2017.
[89] F. Wu, J.-M. Redoute, and M. R. Yuce, ‘‘A self-powered wearable body [109] M. Chen, Y. Ma, Y. Li, D. Wu, Y. Zhang, and C.-H. Youn, ‘‘Wearable
sensor network system for safety applications,’’ in Proc. IEEE Sensors, 2.0: Enabling human-cloud integration in next generation healthcare
New Delhi, India, Oct. 2018, pp. 1–4. systems,’’ IEEE Commun. Mag., vol. 55, no. 1, pp. 54–61, Jan. 2017.
[110] J. Dian and R. Vahidnia, ‘‘Radiation safety hazards of cellular IoT REZA VAHIDNIA (Member, IEEE) was born in
devices,’’ IEEE Technol. Soc. Mag., 2019. Tehran, Iran. He received the B.Sc. and M.Sc.
[111] R. Lomotey, K. Sofranko, and R. Orji, ‘‘Enhancing privacy in wearable degrees in electrical engineering from Tehran
IoT through a provenance architecture,’’ Multimodal Technol. Interact., and Tarbiat Modares University, Tehran, Iran,
vol. 2, no. 2, p. 18, 2018. in 2006 and 2009, respectively, and the Ph.D.
degree from Ontario Tech University, Oshawa,
Canada, in 2014. From 2014 to 2015, he was
a Postdoctoral Fellow with Ryerson University,
where he conducted research on energy harvesting
systems. Since 2016, he has been a Senior Internet
of Things (IoT) Specialist with Rogers Communications Inc., and TELUS.
He is currently a Faculty Member with the Department of Electrical and
Computer Engineering, British Columbia Institute of Technology (BCIT),
Vancouver, Canada. His research interests include wireless communications,
the IoT, and signal processing.